International Journal of Intelligent Engineering and Systems

1

Authors: Balaji Sreenivasulu, Anjaneyulu Pasala, Gaikwad Vasanth

In computer vision, domain adaptation or transfer learning plays an important role because it learns a target classifier characteristics using labeled data from various distribution. The existing researches mostly focused on minimizing the time complexity of neural networks and it effectively worked on low-level features. However, the existing method failed to concentrate on data augmentation time and cost of labeled data. Moreover, machine learning techniques face difficulty to obtain the large amount of distributed label data. In this research study, the pre-trained network called inception layer is fine-tuned with the augmented data. There are two phases present in this study, where the effectiveness of data augmentation for Inception pre-trained networks is investigated in the first phase. The transfer learning approach is used to enhance the results of the first phase and the Support Vector Machine (SVM) is used to learn all the features extracted from inception layers. The experiments are conducted on a publicly available dataset to estimate the effectiveness of proposed method. The results stated that the proposed method achieved 95.23% accuracy, where the existing techniques namely deep neural network and traditional convolutional networks achieved 87.32% and 91.32% accuracy respectively. This validation results proved that the developed method nearly achieved 4-8% improvement in accuracy than existing techniques.

20

Authors: Reza Fuad Rachmadi, I Ketut Eddy Purnama, Supeno Mardi Susiki Nugroho, Yoyon Kusnendar Suprapto

Faces is a unique region in our body that can be used as a biometric identity. Furthermore, the face between two people that have a kinship relationship may share the same face features which can be used to decide whether two people have a kinship relationship or not. In this paper, we proposed a family-aware convolutional neural network (CNN) for the visual kinship verification problem. Our proposed classifier is constructed by paralleling the state-of-the-art face recognition model and attaching two additional networks, a family-aware network, and a kinship verification network. The family-aware network weights adjusted by learning features specific to the family using deep metric learning loss while the kinship verification network use softmax loss to learn the kinship verification problem. One of the advantages of our proposed classifier is that the output of the classifier is normalized and can be represented as the probability of two images being kin or non-kin. To preserve the face recognition features extraction ability in the state-of-the-art face recognition model, we freeze the weights of the convolutional layers in the classifier for the training process. In the testing process, the family-aware network is detached to construct the final classifier. Experiments on FIW (Families In the Wild) dataset show that our proposed classifier performs better comparing with classifiers that trained without a family-aware network and the ensemble version of the classifier is comparable with several state-of-the-art methods with an average accuracy of 68.84%.

42

Authors: Archana Kollu, Sucharita Vadlamudi

Energy management of the cloud datacentre is a challenging task, especially when the cloud server receives a number of the user’s request simultaneously. This requires an efficient method to optimally allocate the resources to the users. Resource allocation in cloud data centers need to be done in optimized manner for conserving energy keeping in view of Service Level Agreement (SLA). We propose, Eagle Strategy (ES) based Modified Particle Swarm Optimization (ES-MPSO) to minimize the energy consumption and SLA violation. The Eagle Strategy method is applied due to its efficient local optimization technique. The Cauchy Mutation method which schedules the task effectively and minimize energy consumption, is applied to the proposed ES-MPSO method for improving the convergence performance. The simulation result shows that the energy consumption of ES-MPSO is 42J and Particle Swarm Optimization (PSO) is 51J. The proposed method ES-MPSO achieves higher efficiency compared to the PSO method in terms of energy management and SLA.

67

Authors: Nageswar Rao Amgoth, Rajendra Naik Bhukya, Nirmala Devi Lavadya

In the field of Wireless Sensor Networks (WSNs), after the deployment of sensor nodes, transmission of data with an effective energy is becoming a critical issue due to the limited battery of sensor nodes. To solve this issue, most of the past developed approaches focused on the implementation of clustering in which the sensor nodes forward their data through Cluster Heads (CHs) to Base Station (BS). However, the main issue is to select an optimal CH or CHs such that every sensor node is ensured with at least one path to base station, directly, through CHs or through relay nodes. Towards this objective, we have proposed a new method called Multi-Objective Assisted Clustered Relay Node Placement (MOCRNP) to achieve improved energy efficiency along with an enhanced network lifespan. Firstly, to select the CH, the Multi-Objective Genetic Algorithm is introduced based on energy, distance and node density. Secondly, in the process of rely node placement, Multi-objective firefly algorithm is introduced and for this purpose, two constraints such as Coverage and Connectivity are considered. We have conducted several experimentations using several setups of the WSN and the performance is evaluated in terms of Network lifetime, average energy consumption, average residual energy, and throughput. On an average, the proposed approach has gained an improvement of 150 sec network lifetime, 104 Kbps Throughput when compared with state-of-art methods.

94

Authors: Rajeev Kumar, Mamdouh Alenezi, Md Tarique Jamal Ansari, Bineet Kumar Gupta, Alka Agrawal, Raees Ahmad Khan

Nowadays, most of the cyber-attacks are initiated by extremely malicious programs known as Malware. Malwares are very vigorous and can penetrate the security of information and communication systems. While there are different techniques available for malware analysis, it becomes challenging to select the most effective approach. In this context, the decision-making process may be an efficient means of empirically assessing the impact of different methods for securing the web applications. In this research study, we have used a methodology that includes the integration of Fuzzy AHP and Fuzzy TOPSIS technique for evaluating the impact of different malware analysis techniques in web application perspective. This study uses different versions of a university’s web application for evaluating the impact of several existing malware analysis techniques. The findings of the study show that the Reverse Engineering approach is the most efficient technique for analyzing complex malware. The outcome of this study would definitely aid the future researchers and developers in selecting the appropriate techniques for scanning the web application code and enhancing the security.

120

Authors: Neelampalli Jayanthi, Burra Vijaya Babu, Nandam Sambasiva Rao

The outlier detection technique is widely used in the data analysis for the clustering of data. Many techniques have been applied in the outlier detection to increase the efficiency of the data analysis. The Local Projection based Outlier Detection (LPOD) method effectively identifies neighbouring values of data, but this has the drawback of random selection of the cluster centre that affects the overall clustering performance of the system. In this study, the Adaptive Clustering by Fast Search and Find of Density Peak (ACFSFDP) is proposed to select the clustering centre and density peak. This ACFSFDP method is implemented with the min-max algorithm to find the number of categories that measured the local density and distance information. The density and distance are used to select the cluster centre, but density is not calculated on the existing distance based clustering techniques. The ACFSFDP method calculates cluster centre based on the density and distance during the clustering process, whereas the existing techniques randomly select the data centre. The results indicated that the ACFSFDP method is provided effective outlier detection compared with existing Clustering by Fast Search and Find of Density Peak (CFSFDP) methods. The ACFSFDP is tested on two datasets Pen-digits and waveform datasets. The experiment results proved that Area Under Curve (AUC) of the ACFSFDP is 99.08% on the Pen-Digit dataset, while the existing distance classifier method k-Nearest Neighbour has achieved 68.7% of AUC.

142

Authors: Pratik Kanani, Mamta Padole

Internet of Things (IoT) generates a myriad amount of data, which is sent over the Cloud computing infrastructure for analytics and Business Intelligence. This application scenario suffers network delays, transmission delays and delays in decision making. Due to these drawbacks, the Cloud-based IoT infrastructure is not suitable for time-critical health care applications. To overcome this problem, a smart way is introduced called “Fog Computing” - a LAN based processing approach which has multiple advantages. When IoT, Fog and Cloud Computing are combined, the resultant system’s performance is far better. Hence, the combination results in a very efficient Health Care system. Fog and Cloud Computing have their dimensions that not only support each other but also explore many new application domains. In this paper, the real-time ElectroCardioGram (ECG) based Health Care system is implemented in Cloud and Fog Computing. Different Quality of Service (QoS) parameters like memory consumption, transmission delays, computation delays, network delays, Carbon dioxide emission, data transferred and response time are measured, analyzed and improved to make the system more efficient. Based on the Fog computing characteristics and capabilities, the Raspberry Pi 3 B+ model is configured as a Health Care serving gateway by using different installation and configuration steps. Initially, the proposed system is tested for one patients ECG data analysis over cloud and Fog. In every set up all QoS parameters are measured and later the system is subjected to multiple ECG streams for varying numbers of patients to find the limitations of the Raspberry Pi node as a Fog Computing node. The obtained results show that for more number of ECG streams the Fog node is not able maintain QoS in decision making time. Every QoS parameter is explored in detail for decision-making time. In the end, the Fog computing based proposed system is concluded for its pros and cons and future aspects of the Fog node are discussed to make better systems.

168

Authors: Pyae Pyae Win Cho, Thi Thi Soe Nyunt

Differential Evolution (DE) has become an advanced, robust, and proficient alternative technique for clustering on account of their population-based stochastic and heuristic search manners. Balancing better the exploitation and exploration power of the DE algorithm is important because this ability influences the performance of the algorithm. Besides, keeping superior solutions for the initial population raises the probability of finding better solutions and the rate of convergence. In this paper, an enhanced DE algorithm is introduced for clustering to offer better cluster solutions with faster convergence. The proposed algorithm performs a modified mutation strategy to improve the DE’s search behavior and exploits Quasi-Opposition-based Learning (QBL) to choose fitter initial solutions. This mutation strategy that uses the best solution as a target solution and applies three differentials contributes to avoiding local optima trap and slow convergence. The QBL based initialization method also contributes to increasing the quality of the clustering results and convergence rate. The experimental analysis was conducted on seven real datasets from the UCI repository to evaluate the performance of the proposed clustering algorithm. The obtained results showed that the proposed algorithm achieves more compact clusters and stable solutions than the competing conventional DE variants. Moreover, the performance of the proposed algorithm was compared with the existing state of the art clustering techniques based on DE. The corresponding results also pointed out that the proposed algorithm is comparable to other DE based clustering approaches in terms of the value of the objective functions. Therefore, the proposed algorithm can be regarded as an efficient clustering tool.

189

Authors: Syahfrizal Tahcfulloh

Unlike the PMIMO radar, the transmit-receive (Tx-Rx) subarrays MIMO (TRSM) radar uses overlaping subarrays in Tx and Rx array so that it simultaneously combines the main advantage of the phased array radar (PA) i.e., high directional coherent gain, and the main advantage of the MIMO radar i.e., high waveform diversity gain. This paper has derived the radar performance formula such that Tx-Rx gain and SINR. The approach aims to overcome the beam shape loss, increase the transmit-receive gain, minimize the maximum peak sidelobe levels, narrow the half power beamwidth, increase directivity, and increase signal-to-noise-plus-interference ratio (SINR). This radar's performance is compared to the PMIMO radar in various methods such as equal subarrays, unequal subarrays, and optimum partitioning, the PA radar, and the MIMO radar. The numerical simulation and evaluation results show that the proposed radar has several advantages such as lowest the peak sidelobe level, narrow the half power beamwidth, and high directivity, so it is very robust against interference effects. When compared to the OPPM radar, as representative of the best-performing radars, this radar has an average performance improvement of MPSLL, directivity, and HPBW which are 21.5 dB, 2 dB, and 0.45 deg, respectively.

208

Authors: Manohar Pundikal, Mallikarjun Sayabanna Holi

The diabetic retinopathy is the leading cause of blindness worldwide, so early detection of diabetic retinopathy is necessary to reduce eye-related diseases. The accurate identification of microaneurysms is crucial for the detection of diabetic retinopathy, because it appears as the first sign of the disease. In this study, a new model is proposed to detect microaneurysms from the retinal images for early diagnosis of diabetic retinopathy. At first, the fundus images are collected from e-ophtha microaneurysms and DiaretDB1 datasets. Next, image pre-processing is accomplished using image normalization, low light image enhancement, gradient weighting and shade correction. The pre-processing methods significantly brighten the contrast of the fundus images for better visual quality and extract the hidden details of the dark conditions. In addition, Hessian based filter and Otsu threshold are used to extract the foreground objects like microaneurysms from the enhanced fundus images. At last, Grey Wolf Optimization (GWO) is used to predict the correctness of segmented microaneurysms candidates. The experimental results have revealed that the proposed model enhanced the microaneurysms detection up to 0.06-0.30 f-score value compared to the other existing models local convergence index features and local features with k-nearest neighbor. In addition, the proposed model has achieved 85.72% and 86.16% of accuracy respectively on e-ophtha microaneurysms and DiaretDB1 datasets.

229

Authors: Nagaraja Rao Pamula Pullaiah, Dorai Venkatasekhar, Padarthi Venkatramana, Balaraj Sudhakar

Breast cancer is one of the most occurring cancers in women due to the uncontrolled growth of abnormal cells in the lobules or milk ducts. The treatment for the breast cancer at an early stage is important using Magnetic Resonance Imaging (MRI) which effectively measures the size of the cancer and also checks tumors in the opposite breast. The deposition of calcium components on the breast tissue is known as micro-calcifications. The calcium salts deposited in the breast are involved with the cancer and were not diagnosed accurately due to the low effectiveness of existing imaging technique namely Haralick feature extraction technique. The MRI breast cancer diagnosis creates problems during classification of breast image and leads to misclassifications, such as unidentified calcium deposits in the existing K-Nearest Neighbour (KNN) classifier. The misclassification issues are overcome by an accurate classification and identification of calcium salts and checks whether deposited salt on breast tissue is involved with cancer or not. Initially, Contrast-Limited Adaptive Histogram Equalization (CLAHE) is used to remove the unwanted noise in the MRI and Morphological, Multilevel Otsu’s Thresholding and region growing techniques perform segmentation to mask unwanted breast tissues. The proposed Hybrid LOOP Haralick feature extraction technique is developed by combining the both Local Optimal Oriented Pattern (LOOP) and Haralick texture feature and the hybrid parameters are applied to the Stacked Auto Encoder based (SAE) to classify the breast MRI image as a Malignant or Benign. The performance of the proposed hybrid LOOP Haralick feature extraction shows significant accuracy improvement of 3.83% when compared to the Haralick feature extraction technique.

255

Authors: Ahmad Afiif Naufal, Chastine Fatichah, Nanik Suciati

This research developed a smart parking system through video data analysis using deep learning techniques that automatically determine the availability of vacant parking spaces. This system has two main stages. The first is the stage of marking the parking position on the image of a parking lot captured by the camera. This research proposes a Preprocessed Region-based Convolutional Neural Network (Mask R-CNN) to mark the parking position on the input image of a full parking lot. The preprocess that combining contrast enhancement using the Exposure Fusion framework, aims to overcome the problem of lighting variations in images taken in an open area. In the second stage, each parking position is examined whether the position is vacant or not using mAlexNet. A series of trials on images with varying light conditions indicate that the Preprocessed Mask R-CNN can improve marking the parking positions with an accuracy of Intersection over Union (IoU) reach 85.80%. The result of marking the parking position is then used in the trial of the availability of parking space on video data using mAlexNet, and achieving an accuracy of 73.73%.

279

Authors: Hanan A. R. Akkar, Sameem Abbas Salman

This paper proposes a new meta-heuristic swarm optimization algorithm called Cicada Swarm Optimization (CISO) algorithm, which mimics the behavior of bio-inspired swarm optimization methods. The CISO algorithm is tested with 23 benchmark functions and taken two problems engineering design, pressure vessel problem and himmelblau’s problem. The performance of CISO algorithm is compared with meta-heuristic well-known and recently proposed algorithms (Cockroach Swarm Optimization (CSO), Grasshopper Optimization algorithm (GOA) and Particle Swarm Optimization (PSO)). The obtained results showed that the proposed algorithm succeeded in improving the test functions and solved engineering design problems that could not be improved by other algorithms according to the chosen parameters and the limits of the research space, also showed that CISO has a faster convergence with the minimum number of iterations and also have an accurate calculation efficiency Satisfactory compared to other optimization algorithms.

307

Authors: Hoai Phong Nguyen, Dinh Tuyen Nguyen, Minh Thuyen Chau

In this paper, a novel approach to realize the carrier-based pulse width modulation (CBPWM) method for indirect matrix converter (IMC) is presented. The advantage of the proposed CBPWM method is that only one triangular carrier signal with constant gradient falling and rising edges is used to generate the PWM to the power switches in both the rectifier and the inverter stages of IMC. The analysis of the proposed CBPWM method is based on the space vector approach and the relationship between CBPWM and space vector pulse width modulation (SVPWM) methods is provided. In this paper, four CBPWM methods called: sinusoidal PWM, third harmonic injection PWM, symmetrical PWM and discontinuous PWM are presented. These different CBPWM schemes are established according to the distribution of zero vectors and the switching pattern of the SVPWM method. The comparison of output voltage performance with different CBPWM schemes are investigated. Compared to the conventional method the calculation time of the proposed method is reduced around 50%. Concretely, the calculation time of the SVPWM method is 45 μs, while the calculation time of the proposed method is 22 μs. The experiments have been carried out to show the effectiveness of the proposed method.

330

Authors: Edi Jaya Kusuma, Guruh Fajar Shidik, Ricardus Anggi Pramunendar

Classification is one of the data mining techniques which considered as supervised learning. Classification technique such as Backpropagation Neural Network (BPNN) has been utilized in several fields to increase human productivity. BPNN can give better results (more natural) compared with other statistical techniques. However, the learning process of BPNN could give an inefficient synapse weight of each hidden layer. This ineffective weight can affect the performance of the network. In this research, BPNN optimization using Nelder Mead to identifying the appearance of breast cancer is proposed. The datasets used are Breast Cancer Coimbra Dataset (BCCD), and Wisconsin Breast Cancer Dataset (WBCD). The testing result using accuracy and k-fold validation presents better performance compared with the original BPNN. Best average performance can be seen in the fifth fold of BCCD with 76.5217% of accuracy. Moreover, the highest average result of WBCD presented in the fourth fold with 91.1765% of average accuracy.

349

Authors: I Wayan Agus Surya Darma, Nanik Suciati, Daniel Siahaan

The preservation of Balinese carving data is a challenge in recognition of Balinese carving. Balinese carvings are a cultural heritage found in traditional buildings in Bali. The collection of Balinese carving images from public images can be a solution for preserving cultural heritage. However, the lousy quality of taking photographs, e.g., skewed shots, can affect the recognition results. Research on the Balinese carving recognition has existed but only recognizes a predetermined image. We proposed a Neural Style Geometric Transformation (NSGT) as a data augmentation technique for Balinese carvings recognition. NSGT is combining Neural Style Transfers and Geometric Transformations for a small dataset solution. This method provides variations in color, lighting, rotation, rescale, zoom, and the size of the training dataset, to improve recognition performance. We use MobileNet as a feature extractor because it has a small number of parameters, which makes it suitable to be applied on mobile devices. Eight scenarios were tested based on image styles and geometric transformations to get the best results. Based on the results, the proposed method can improve accuracy by up to 16.2%.

374

Authors: Vivaram Veera Raghavulu, Ande Prasad

Video analytics is widely used to automatically analyze the videos to extract the required information and detect the various events object identification and traffic analysis. The segmentation of the image is referring to extracting the required region from an image. The major objective of the segmentation process is to cluster the images without being affected by the noises. The detection of the moving objects is a challenging task in video analysis due to the dynamic background of the video. The major drawback of the existing Kernel Fuzzy C-Means (KFCM) clustering is initialization of random centroids which increases the execution time to identify the segmented portions. In this research, the Grey Wolf Optimization (GWO) algorithm used to initialize the centroids of required clusters in KFCM and Level Set (LS) Algorithm is used to segment the objects in video sequence. The proposed KFCM-GWO-LS is implemented for moving and static object detection in the videos obtained from SBM-RGBD dataset. For object detection, determining central clusters are important which is performed by using KFCM. GWO helps in finding the best centroids clusters by matching with KFCM. The centroids clusters are segmented by using LS algorithm which undergoes over segmentation problem that is overcome by GWO. As all the three techniques are dependent on one another hybrid of all these techniques obtains better results. The proposed KFCM-GWO-LS is evaluated in terms of Recall, Specificity, Precision and F-measure and the experimental results showed that the proposed method improved the system performance from 0.3 % to 14.35 % compared to existing methods Multi-Sensor Scheduler Algorithm and Statistical Inference Theory model.

393

Authors: Hossam A. Nabwey

This work reports how to utilize rough set theory and group method analysis for generating a set of rules to investigate heat and mass transfer of mixed convection stagnation point flow of a non-Newtonian nanofluid towards a vertical stretching surface. By utilizing group method analysis, the main partial differential equations which describe the flow are rehabilitated to nonlinear Ordinary Differential Equations (ODEs). Then, the resultant nonlinear ODEs are solved numerically by applying the implicit finite-difference scheme. The numerical values thus obtained are depicted in tabular form and the basic principles of rough sets are applied to get all reducts and finally a group of generalized rules are extracted to predict the value of local Nusselt number and local skin-friction coefficient. The outcomes demonstrate the novelty of the current work which can be summarized as hybridization of group method analysis and rough set theory to use in the field of fluid dynamics effectively. The resultant set of generalized classification rules which performed with basic logic functions can be considered as knowledge base with high accuracy and may be valuable in many engineering applications like power production, thermal extrusion systems and microelectronics.

419

Authors: Irya Wisnubhadra, Safiza Suhana Kamal Baharin, Nanna Suryana Herman

Business Intelligence (BI) technology with Extract, Transform, and Loading process, Data Warehouse, and OLAP have demonstrated the ability of information and knowledge generation for supporting decision making. In the last decade, the advancement of the Web 2.0 technology is improving the accessibility of web of data across the cloud. Linked Open Data, Linked Open Statistical Data, and Open Government Data is increasing massively, creating a more significant computer-recognizable data available for sharing. In agricultural production analytics, data resources with high availability and accessibility is a primary requirement. However, today’s data accessibility for production analytics is limited in the 2 or 3-stars open data format and rarely has attributes for spatiotemporal analytics. The new data warehouse concept has a new approach to combine the openness of data resources with mobility or spatiotemporal data in nature. This new approach could help the decision-makers to use external data to make a crucial decision more intuitive and flexible. This paper proposed the development of a spatiotemporal data warehouse with an integration process using service-oriented architecture and open data sources. The data sources are originating from the Village and Rural Area Information System (SIDeKa) that capture the agricultural production transaction in a daily manner. This paper also describes the way to spatiotemporal analytics for agricultural production using a new spatiotemporal data warehouse approach. The experiment results, by executing six relevant spatiotemporal query samples on DW with fact table contains 324096 tuples with temporal integer/float for each tuple, 4495 tuples of field dimension with geographic data as polygons, 80 tuples of village dimension, dozens of tuples of the district, subdistrict, province dimensions. The DW time dimension contains 3653 tuples representing a date for ten years, proved that this new approach has a convenient, simple model, and expressive performance for supporting executive to make decisions on agriculture production analytics based on spatiotemporal data. This research also underlines the prospects for scaling and nurturing the spatiotemporal data warehouse initiative.

442

Authors: Rudy Dikairono, Setiawardhana, Djoko Purwanto, Tri Arief Sardjono

The Convolutional Neural Network (CNN) is an object classification method that has been widely used in recent research. In this paper, we propose CNN for use in the self-localization of wheeled soccer robots on a soccer field. If the soccer field is divided into equally sized quadrants with imaginary vertical and horizontal lines intersecting in the middle of the field, then the soccer field has an identical shape for each quadrant. Every quadrant is a reflection of the other quadrants. Superficially similar images appearing in different positions may result in positioning mistakes. This paper proposes a solution to this problem by using a visual modelling of the gyrocompass line mark and omni-vision image for the CNN-based self-localization system. A gyrocompass is used to obtain the angle of the robot on the soccer field. A 360° omni-vision camera is used to capture images that cover all parts of the soccer field wherever the robot is located. The angle of the robot is added to the omni-vision image using the visual modelling method. The implementation of self-localization without visual modelling gives accuracy rates of 0.3262, and this result is increased to 0.6827 with the proposed methods. The experiment was carried out in the robotics laboratory of the Institut Teknologi Sepuluh Nopember (ITS) with the ITS Robot with Intelligent System (IRIS) robot.

460

Authors: Ahmed Rafid Hashim, Hazim G. Daway, Hana H. kareem

Haze causes the degradation of image quality. Thus, the quality of the haze must be estimated. In this paper, we introduce a new method for measuring the quality of haze images using a no-reference scale depending on color saturation. We calculate the probability for a saturation component. This work also includes a subjective study for measuring image quality using human perception. The proposed method is compared with other methods as, entropy, Naturalness Image Quality Evaluator (NIQE), Haze Distribution Map based Haze Assessment (HDMHA), and no reference image quality assessment by using Transmission Component Estimation (TCE). This done by calculating the correlation coefficient between non-reference measures and subjective measure, the results show that the proposed method has a high correlation coefficient values for Pearson correlation coefficient (0.8923), Kendall (0.7170), and Spearman correlation coefficient (0.8960). The image database used in this work consists of 70 hazy images captured by using a special device, design to capture haze image. The experiment on haze database is consistent with the subjective experiment.

487

Authors: Hanan A. R. Akkar, Suhad Qasim G. Haddad

The most significant challenge facing the researcher in the field of robotics is to control the robot manipulator with appropriate overall performance. This paper focuses mainly on the novel Intelligent Particle Swarm Optimization (PSO) algorithm that was used for optimizing and tuning the gain of conventional Proportional Integral Derivative (PID), and improve the parameters of dynamic design in Sliding Mode Control (SMC), which is considered a strong nonlinear controller for controlling highly nonlinear systems, particularly for multi-degree serial link robot manipulator. Additional modified Integral Sliding Mode Controller (ISMC) was implemented to the design of dynamic system with high control theory of sliding mode controller. Intelligent Particle Swarm Optimization (PSO) algorithm was introduced for developing the nonlinear controller. The algorithm demonstrates superior performance in determining the appropriate gains and parameters value in harmony with robot scheme dynamic layout in order to achieve suitable and stable nonlinear controller, besides reduce the chattering phenomenon. PUMA robot manipulator that was used as study case in this work, shows perfect result in step response, with acceptable steady state, and overshoot, besides, eliminating the disadvantage of chattering in conventional SMC. Matlab / Simulink presents to increase the speed of matrix calculation in forward, inverse kinematics and dynamic model of manipulator. Comparison was made between the proposed method with existing methods. Result shows that integral sliding mode with PSO (ISMC/PSO) gave best result for stable step response, minimum mean square error with best objective function, and stable torque.

512

Authors: Mohamed Taha, Mohamed Ibrahim, Hala H. Zayed

Vein detection is an important issue for the medical field. There are some commercial devices for detecting veins using infrared radiation. However, most of these commercial solutions are cost-prohibitive. Recently, veins detection has attracted much attention from research teams. The main focus is on developing real-time systems with low-cost hardware. Systems developed to reduce costs suffer from low frame rates. This, in turn, makes these systems not suitable for real-world applications. On the other hand, systems that use powerful processors to produce high frame rates suffer from high costs and a lack of mobility. In this paper, a real-time vein mapping prototype using augmented reality is proposed. The proposed prototype provides a compromised solution to produce high frame rates with a low-cost system. It consists of a USB camera attached to an Android smartphone used for real-time detection. Infrared radiation is employed to differentiate the veins using 20 Infrared Light Emitting Diodes (LEDs). The captured frames are processed to enhance vein detection using light computational algorithms to improve real-time processing and increase frame rate. Finally, the enhanced view of veins appears on the smartphone screen. Portability and economic cost are taken into consideration while developing the proposed prototype. The proposed prototype is tested with people of different ages and gender, as well as using mobile devices of different specifications. The results show a high vein detection rate and a high frame rate compared to other existing systems.

533

Authors: Nuha H. Abdulghafoor, Hadeel N. Abdullah

Multi-object detection and tracking systems represent one of the basic and important tasks of surveillance and video traffic systems. Recently. The proposed tracking algorithms focused on the detection mechanism. It showed significant improvement in performance in the field of computer vision. Though. It faced many challenges and problems, such as many blockages and segmentation of paths, in addition to the increasing number of identification keys and false-positive paths. In this work, an algorithm was proposed that integrates information on appearance and visibility features to improve the tracker's performance. It enables us to track multiple objects throughout the video and for a longer period of clogging and buffer a number of ID switches. An effective and accurate data set, tools, and metrics were also used to measure the efficiency of the proposed algorithm. The experimental results show the great improvement in the performance of the tracker, with high accuracy of more than 65%, which achieves competitive performance with the existing algorithms.

560

Authors: Ike Fibriani, Widjonarko, Aris Prasetyo, Angga Mardro Raharjo, Dasapta Erwin Irawan

The COVID-19 pandemic has become the focus of world problems that need to be resolved. This is because the rate of spread is speedy and able to take down the world's health system. Therefore, many researchers are focusing their research on solving this problem by doing an initial screening on the X-Ray image of the subject's lungs. One of them is by using Deep Learning. Several articles that talk about implemented Deep Learning for classifying X-Ray images have been published. But most of them are comparing different architecture CNN (Convolutional Neural Network). In this study, the authors try to create a multi-classifier Deep Learning system that consists of nine different CNN architectures and combined with three different Majority Vote techniques. The target of this research is to maximize the performance of classification and to minimize errors because the final decision is a compilation of decisions contained in each CNN architecture. Several models of CNN are tested in this study, both the model which used Majority Vote and Conventional CNN. The results show that the proposed model achieves an accuracy value average F1-Score 0.992 and Accuracy 0.993, according to 5 K-Fold test. The best model is CNN, which used Soft Majority Vote.